An advantage of the bending energy approach is that it minimizes the chance of collapsing landmarks because changes in the positions of landmarks that are close together requires much more bending energy. This is not usually a problem but it can be if the outline has a sharp corner - which seems likely for a beak.
---------------------- F. James Rohlf New email: f.james.ro...@stonybrook.edu Distinguished Professor, Emeritus. Dept. of Ecol. & Evol. & Research Professor. Dept. of Anthropology Stony Brook University 11794-4364 The much revised 4th editions of Biometry and Statistical Tables are now available: http://www.whfreeman.com/Catalog/product/biometry-fourthedition-sokal http://www.whfreeman.com/Catalog/product/statisticaltables-fourthedition-rohlf Please consider the environment before printing this email -----Original Message----- From: Guillermo Navalón [mailto:guiyelm...@gmail.com] Sent: Monday, August 22, 2016 9:34 AM To: MORPHMET <morphmet@morphometrics.org> Subject: [MORPHMET] Problems with min-dsquare sliding in tpsRelw Hi everyone, In a very disparate sample of bird skulls I am using a configuration with both lnmdks and smlndmks. Specifically, to capture the lateral morphology of the beak (likely the most variable area) I digitized 2 curves with 15 evenly-spaced semilandmarks. The 2 curves are constrained by 3 regular lndmks forming a triangle, the tip-of-the-beak landmark (landmark 1) is the anterior end of both curves. When I slide the smlndmks in tpsRelw with min-dsquare slide method some of the anterior smlndmks collapse in a very narrow section in both of the 2 curves in the Procrustes superimposition. This effect is not affecting to the other curve in the configuration (midline of the neurocranium) that is apparently much less variable in my sample. Also, minimum bending energy slide method does not affect the superimposition in this way, but I want to try to use min-dsquare. I have tried to change the slide maximum iterations and slide recursive options but still recover the same effect. Any idea what is going on here? Is it a bug of the program or is it a proper statistical effect? Thank you! Guillermo Navalón -- MORPHMET may be accessed via its webpage at http://www.morphometrics.org --- You received this message because you are subscribed to the Google Groups "MORPHMET" group. To unsubscribe from this group and stop receiving emails from it, send an email to morphmet+unsubscr...@morphometrics.org. -- MORPHMET may be accessed via its webpage at http://www.morphometrics.org --- You received this message because you are subscribed to the Google Groups "MORPHMET" group. To unsubscribe from this group and stop receiving emails from it, send an email to morphmet+unsubscr...@morphometrics.org.
